The clinical literature suggests that intermittent cycles of repeated cocaine dosing, i.e., binges, followed by abrupt cessation of cocaine use are associated with a distinct pattern of withdrawal symptoms and signs, with the probability of relapse varying across phases of cocaine abstinence. A significant hindrance to the understanding of the relationship between cocaine abstinence and treatment responsivity is the lack of laboratory studies in which cocaine binge behavior is manipulated, cocaine access is terminated, and withdrawal sequelae elucidated. In the proposed research, experienced male and female cocaine users residing in a controlled setting, will be given the opportunity to self- administer cocaine by their laboratory. Subjective, performance and physiological changes will be studied during prolonged cocaine abstinence. The proposal is designed to specifically investigate the following questions: 1) Can changes in affect, including depressive symptomatology, and sleep, biological markers of neurotransmitter activity, and reports of withdrawal symptoms following the termination of a cocaine binge be related to the magnitude and duration of the preceding cocaine binge? 2) How will subjects' reports of cocaine "craving" change during cocaine abstinence, and will this pattern vary in response to a) time, b) classically conditioned cocaine-cues, and c) doses of cocaine after abstinence? 3) What are the difference/similarities in response to cocaine abstinence following binges when cocaine is smoked compared to binges when cocaine is taken i.v.? 4) Will pharmacological interventions be useful in ameliorating cocaine withdrawal symptomatology? These studies will provide information relevant to Gawin & Kleber's model of cocaine dependence, and have immediate clinical relevance. An analysis of cocaine abstinence will prove data necessary to understand behavioral changes in subjects starting treatment programs, suggesting factors that may be critical to patient compliance. The development of this laboratory model of abstinence will also provide a starting point for the development of laboratory models of abstinence phenomena with other drugs of abuse. Such models should play a role in understanding initial abstinence and how these factors, including conditioned cues, relate to relapse.